The present invention relates to computer systems management, and more specifically, to monitoring parameters of an operating system for deviance from a baseline behavior and to triggering diagnostic data gathering in response to the deviance exceeding a predetermined threshold.
When contemporary operating systems experience an error or unexpected result in a computing environment, a great deal of time and resources are typically spent determining the root cause of the error or unexpected result. Typically, insufficient documentation is collected on the operating systems to allow a complete diagnosis to be performed. Clients and testers are required to recreate error situations to attempt to gather additional documentation on a subsequent failure. By the time a software problem percolates up to the operating system, making itself known, the initial problem may have already disappeared from traces, making the debugging of the initial problem difficult.